JP2010028875A - Spread spectrum signal processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for processing spread spectrum signals. <P>SOLUTION: A continuous signal of a comparatively high frequency is received (1610). This signal is sampled at a basic sampling rate whereby a resulting sequence of time-discrete signal samples is provided, which are in turn quantized into a corresponding level-discrete sample value (1620). A plurality of data words are formed, which each includes one or more consecutive sample values (1630). Information obtained from these data words is correlated with at least one representation of a signal source specific code sequence (1640), which has been pre-generated in the form of a code vector (1600). The correlation step specifically involves correlating at least each vector in a sub-group of the code vectors with at least one vector that has been derived from the data word. Thereby resulting data are produced (1650). <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates generally to the processing of spread spectrum signals according to a vector-based algorithm, and more particularly, the invention relates to a method for receiving a spread spectrum signal according to the preamble of claim 1 and The present invention relates to a signal receiver according to the preamble of claim 28, and the present invention relates to a computer program according to claim 26 and a computer-readable recording medium according to claim 27.

  Conventionally, spread spectrum technology has become increasingly important in many areas, and mobile communication systems and global navigation satellite systems (GNSS) fall into two important examples. In addition, many transmission standards create demand for hybrid or multipurpose receivers. For example, the mobile phone standards cdma2000 and WCDMA require transmission of a spread spectrum signal. However, due to differences in the air interface, the same terminal cannot be used in two systems. On the contrary, dedicated terminals must be used in each system. Alternatively, a dual mode terminal must be designed, which comprises two separate transceiver chains. In contrast, navigation receivers adapted to certain GNSSs, such as the global positioning system (GPS), can be found in Galileo systems (European programs for global navigation services) or global It is not possible to receive signals from satellites belonging to different GNSSs, such as the global orbiting navigation satellite system (GLONASS).

  To make this possible, a multimode receiver must be designed. However, providing a multi-mode receiver chain in a single device not only increases costs, but also makes the unit larger and heavier, especially if more than two signals must be processed. Therefore, a programmable software receiver solution is desired, and its signal processing scheme will vary depending on the signal received and processed.

  Various software solutions for processing GNSS signals are already known. Patent document WO02 / 50561 is an example in which a GPS receiver tracking system for guiding missiles is described. The receiving station here comprises one or more processing stages for receiving and processing GPS data. The control signal controls whether the number of processing stages should be increased or decreased in response to the result of the interaction. Timing information recovery and tracking is fully implemented in the software. However, the process requires a relatively complex real-time calculation of the Fourier transform and thus requires a large amount of processing resources. Thus, this type of receiver is not particularly suitable for small sized units, such as mobile phones or mobile GNSS receivers.

  The paper “TUPM 12.4: Software solution of GPS Baseband Processing”, ULSI Laboratories, Mitsubishi Electric Corporation (Registered Trademark), IEEE, 1998, Arai T, etc. are for software GPS receivers using embedded microprocessors. The principle is outlined, which can demodulate signals from up to 8 satellites in about 40 MIPS (millions of or or Mega Instructions Per Second). The 2 MHz GPS baseband signal is now fed to a software radio receiver, which is not only phase compensation, but also spectrum despreading and mixing; synchronous demodulation, satellite Perform satellite acquisition and tracking. An algorithm is proposed whereby an intermediate frequency (IF) signal is sampled 1 bit where downconversion is performed prior to despreading. The result data in 32 sample formats are subsequently processed in parallel. A DRAM (Dynamic Random Access Memory) stores a specific C / A (Coarse Acquisition) code table for each satellite to which a signal is transmitted. All other data and instructions are stored in cache memory for high speed processing. Although this strategy is recognized as providing a reasonably effective implementation, this paper lacks a specific description of how despreading and association actually work to achieve this effect.

  Akos.D et al., “Tuning In to GPS-Real-Time Software Radio Architecture for GPS Receivers”, GPS World, July 2001, describes the receiver architecture, through which the IF signal samples are wireless front-end. (radio front-end) to the programmable processor directly for continuous processing. This architecture describes the possibility of using a single instruction multiple data (SIMD) instruction to process multiple data samples in parallel. However, there is no teaching on how effective parallel processing can actually be performed.

  "Design and Test-Bed Implementation of a Reconfigurable Receivable for Nigation" And the receiver can fusing data from two or more different GNSSs. This paper outlines an architecture with a field programmable gate array (FPGA) and a digital signal processor (DSP) in addition to a wireless front end. The authors are working on a variety of computational load problems and they do not present an algorithm that meets the identified requirements.

  Therefore, the prior art includes many examples of software-based GNSS receivers. Nonetheless, it is a highly effective solution that is suitable for software implementation and has performance comparable to that of modern ASIC-based solutions for receiving and processing spread spectrum signals at least in real time. There is currently no clear teaching of the strategy (ASIC = Application Specific Integrated Circuit).

  Accordingly, it is an object of the present invention to provide a solution for receiving and processing a spread spectrum signal, which solves the above problems and is therefore implemented in a physically small and power efficient device. Present a truly software conformance strategy.

  According to one aspect of the present invention, this object is achieved by a method for receiving a spread spectrum signal as described at the outset, characterized in that it comprises a preprocessing step of association, said preprocessing step Is performed before reception of a continuous signal. This preprocessing includes a number of code vector pre-generation steps, each of which represents a specific code sequence of at least one signal source specific code sequence. Further in accordance with the present invention, the associating step includes multiplying each vector in at least a subgroup of code vectors by at least one vector obtained from the data word.

  This strategy is advantageous. Because, the proposed vector approach allows a digital processor (eg, a microprocessor) to process multiple signal samples in parallel during each clock cycle, thus making the use of the processor very efficient. Further, pre-generating code vectors saves valuable processing power, which is a sufficient amount of time units to continue to track navigation signals transmitted by hygiene, for example, in navigation satellite systems. Allows processing of per-input data.

  According to a preferred embodiment of this aspect of the invention, each code vector represents a particular signal source specific code sequence, which is sampled at a basic sampling rate and generates level discrete sample values. It is quantized using the quantization process used. Preferably, the code vector is further adapted to the format of the input data signal in a way that makes the following processing efficient.

  According to another preferred embodiment of this aspect of the invention, the at least one signal source specific code sequence represents so-called pseudo random noise. This type of code sequence is useful in an environment where two or more different signals are modulated at the same carrier frequency. That is, the different pseudorandom noise signals are orthogonal (or at least almost orthogonal) to each other. Thus, the first information signal spread by the first pseudo-random noise signal causes a minimum degradation of the second information signal spread by the second pseudo-random noise signal, and vice versa.

  According to a first preferred embodiment of this aspect of the invention, the receiving step includes down conversion from an input high frequency signal to an intermediate frequency signal. The high frequency signal is estimated to have a spectrum that is symmetric around the first frequency, and the intermediate frequency signal is estimated to have a spectrum that is symmetric around the second frequency, which is Much lower than the frequency. Thus, the receiving step converts the information carrier signal into a baseband that is technically easier to handle than the band in which the original input signal is located.

  According to another preferred embodiment of this aspect of the invention, the method comprises the following steps. First, the maximum frequency variation of the second frequency due to the Doppler effect is determined. Then, a Doppler frequency interval near the second frequency is defined. This Doppler frequency interval is the lowest frequency limit equal to the difference between the second frequency and the maximum frequency variation, and the highest frequency equal to the sum of the second frequency and the maximum frequency variation. Limit (highest frequency limit). The Doppler frequency interval is then divided into an integer number of equidistant frequency steps, and then a frequency candidate vector is defined for each frequency step. This method of frequency division is advantageous. This is because it provides flexible modeling of intermediate carrier frequencies and variations thereof.

  According to another preferred embodiment of this aspect of the invention, an integer number of initial phase positions are determined for the frequency candidate vectors. Thus, this integer number represents a phase resolution such that a relatively high number corresponds to a relatively high phase resolution, while a relatively low number corresponds to a relatively low phase resolution. Further, a carrier frequency-phase candidate vector is defined for each combination of a carrier frequency candidate vector and an initial phase position. Different frequency phase candidate vectors are beneficial because they allow the determination of a very accurate estimate of the frequency and phase characteristics of the received signal, which in turn ensures a demodulated signal with high quality.

  According to another preferred embodiment of this aspect of the invention, the number of elements in each carrier frequency phase candidate vector is determined. Subsequently, the carrier frequency phase candidate vector is stored according to the data format, which is adapted to perform a multiplication of the carrier frequency phase candidate vector segment and the data word. This adaptation is advantageous. This is because a large number of signal samples are processed in parallel by a relatively simple operation. Preferably, the adaptation of the data format is such that at least one of the elements of the carrier frequency phase candidate vector is obtained such that the segment acquires a number of elements equal to the number of elements in each of the at least one vector obtained from the data word. Adding to the segment. Thus, it is possible to process a segment of the carrier frequency phase candidate vector with one of the at least one vector either by so-called SIMD (single instruction multiple data) operation or logical XOR operation.

  According to yet another preferred embodiment of this aspect of the invention, the method also includes the following steps. First, the maximum code rate variation due to the Doppler effect is determined. A Doppler rate interval is defined around the center code rate. The Doppler frequency interval has a minimum code rate limit equal to the difference between the center code rate and the maximum code rate variation, and a maximum frequency limit equal to the sum of the center code rate and the maximum code rate variation. Similar to the intermediate frequency spectrum, the Doppler rate interval is divided into an integer number of equidistant code rate steps, and code rate candidates are defined for each code rate step. Further, as with the intermediate frequency, an integer number of possible initial code phase positions are determined for each code rate candidate. Thus, this integer number represents a code phase resolution such that a relatively high number represents a relatively high code phase resolution while a relatively low number represents a relatively low code phase resolution. Typically, spread spectrum code modulation is less sensitive to distortion due to the Doppler effect than frequency modulation. Accordingly, the code rate step may be relatively large. However, an estimate of some Doppler shift is desirable to obtain a high quality modulated signal.

  According to yet another preferred embodiment of this aspect of the invention, a set of code vectors is generated for each source-specific code sequence by sampling each code rate phase candidate vector at a basic sampling rate, and thus A corresponding code vector is generated. Again, this simplifies parallel processing of a large number of signal samples with relatively simple operations.

  According to yet another preferred embodiment of this aspect of the invention, the modified code vector copies a certain number of elements from the end of the original code vector to the beginning of the modified code vector, and the original It is generated based on each code vector by copying a specific number of elements from the beginning of the code vector to the end of the code vector. This extension of the code vector is very advantageous. Because it extends the tracking of parameters through early-, prompt-, and late techniques by performing a simple transformation of a local copy of the source-specific code sequence Enable.

  According to yet another preferred embodiment of this aspect of the invention, a set of modified code vectors is stored for each signal source specific code sequence. Here, each modified code vector includes a number of elements that represent a sampled version of at least one complete code vector. As described above, the modified code vector also includes a repetition at the beginning and end of the sequence. A particular modified code vector is defined for each combination of code rate candidate and code phase position. These pre-generated correction code vectors do not need to be calculated in real time, but are simply obtained from memory, saving considerable processing in the processor.

  According to yet another preferred embodiment of this aspect of the invention, the data format of the modified code vector is such that the modified code vector and one of the at least one vector are combined together by SIMD or XOR operations. As processed, it is adapted with respect to the data format of at least one vector obtained from the data word. Inevitably, this is advantageous in terms of processing efficiency.

  According to yet another preferred embodiment of this aspect of the invention, the method includes an initial acquisition phase and a subsequent tracking phase. The acquisition phase establishes a set of preliminary parameters needed to trigger the decoding of signals received during the tracking phase. This parameter includes the modified code vector, the carrier frequency candidate vector, the initial phase position, the code phase position, and the code index, which means the starting sample value for the modified code vector. Thus, a successful acquisition phase results in that at least one source-specific code sequence is identified and results in that this signal can be tracked (ie continuously received) by the receiver. Bring.

  According to yet another preferred embodiment of this aspect of the invention, the step of tracking in sequence includes the following steps. First, based on the tracking characteristics, a prompt pointer is calculated for each modified code vector. The prompt pointer points to the start position of the code sequence. The initial prompt pointer value is set equal to the code index. At least a pair of early- and late- pointers are assigned near each prompt pointer. The early pointer identifies the sample value placed in at least one element before the prompt pointer location, while the late pointer identifies the sample value placed in at least one element after the prompt pointer location To do. These pointers are placed with the prompt pointer as close as possible to the correlation maximum value, the early pointer is placed in time slightly earlier than this value, and the late pointer is above this value. Is used to maintain tracking of the received signal by iteratively repositioning the pointers near the correlation maximum so that they are placed in time later. An important advantage gained by this strategy is that the same vector can be reused to represent different delays. Thus, valuable memory space and processing power is saved.

  According to yet another preferred embodiment of this aspect of the invention, the tracking step further includes the following steps. First, a series of incoming level discrete sample values is received. Each data word is then formed from sample values such that each data word includes a number of elements equal to the number of elements in each carrier frequency phase candidate vector. A corresponding set of carrier frequency phase candidate vectors is then computed for the data word, and a pre-generated quadrature-phase representation vector of each in the set is generated. Obtained for a carrier frequency phase candidate vector. Subsequently, each data word is, on the one hand, multiplied with an in-phase representation of the carrier frequency phase candidate vector in the corresponding set to produce a first intermediate-frequency-reduced information word, and on the other hand , Multiplying by a quadrature-phase representation of the carrier frequency phase candidate vectors in the corresponding set to generate a second intermediate frequency reduced information word. Thus, the intermediate frequency reduced information word constitutes each element of the above-mentioned vector obtained from the incoming data word. This multiplication strategy is advantageous because it allows a very efficient use of the performance of a general purpose microprocessor.

  In accordance with yet another preferred embodiment of this aspect of the invention, the XOR or SIMD operation includes a multiplication between the in-phase representation of the carrier frequency phase candidate vector and the data word, an orthogonal phase representation of the carrier frequency phase candidate vector, and Used to perform each multiplication with a data word.

  According to another preferred embodiment of this aspect of the invention, the tracking step multiplies the modified code vector starting from the position pointed to by the prompt pointer and the first intermediate frequency reduced information word. Generating a first prompt-despread symbol string and associating the first intermediate frequency reduced information word with the modified code vector starting at the position pointed to by the early pointer Generating a first early despread symbol string and associating the first intermediate frequency reduced information word with the modified code vector starting at the position pointed to by the late pointer Produce despread symbol string and prompt second intermediate frequency reduced information word Generating a second prompt despread symbol string in association with the modified code vector starting at the position pointed to by the pointer, and the second intermediate frequency reduced information word at the position pointed to by the early pointer To generate a second early despread symbol string associated with the modified code vector starting at, and modifying the second intermediate frequency reduced information word starting at the position pointed to by the late pointer Generating a second late despread symbol string in association with the generated code vector. Preferably, the resulting data word is then obtained for each set of despread symbol strings. This can be done, for example, by performing the appropriate addition operation, or depending on whether the resulting data word contains packed or un-packed information. This is done by looking up a pre-generated value in a table based on the bit pattern of Thus, high quality data can be obtained with minimal real-time calculations, which is advantageous from a processing point of view.

  According to yet another preferred embodiment of this aspect of the invention, either an XOR operation or a SIMD operation is also used to perform a multiplication of the intermediate frequency reduced information word and the modified code vector. Again, this is advantageous from a processing point of view.

  According to another preferred embodiment of this aspect of the invention, a portion of information is propagated in connection with completing processing of the current data word and initiating processing of the next data word. (propagate) Preferably, the propagated information includes a pointer indicating the first sample value of the next data word, a group of parameters describing a corresponding set of carrier frequency phase candidate vectors, a corresponding set of code vectors, a prompt, Indicate early and late pointer. Thus, the next data word is immediately processed based on the propagated information.

  According to a further aspect of the present invention, this object is achieved by a computer program that can be loaded directly into the memory of a computer, which is the method according to the above proposal when said program runs on the computer. The software for executing is provided.

  According to another aspect of the present invention, this object is achieved by a computer-readable recording medium, which comprises a program recorded on the recording medium, the program according to the above proposal on a computer. This is to execute the method.

  According to another aspect of the invention, this object is achieved by a signal receiver as described above, wherein the signal receiver is a computer program according to the proposal so that the receiver operates according to the method. It is characterized by being loaded into the memory means.

  Thus, the present invention provides an excellent means for receiving and decoding any type of spread spectrum signal in a general software defined radio receiver. For example, the same receiver can be used to track different types of navigation satellite signals. These signals may be received in parallel, such that the first signal is received in the first format while one or more other signals are received in the second or third format. Good.

Furthermore, the solution according to the above proposal allows the same receiver to be used for completely different purposes, such as communication in a mobile phone system.
The invention will be described in more detail by means of preferred embodiments with reference to the accompanying drawings, which are disclosed by way of example only.

FIG. 2 is a spectrum diagram on a representative spread spectrum signal from which information according to the present invention is extracted. FIG. 2 is a diagram showing a spectrum graph on a frequency down-converted signal corresponding to the spectrum shown in FIG. 1. FIG. 3 is an enlarged view of FIG. 2 illustrating the Doppler shift of the spectrum. FIG. 6 illustrates how a data signal according to an embodiment of the present invention is modulated into a signal source specific code sequence. FIG. 6 shows a three-dimensional graph for a carrier frequency phase candidate vector to be used as an initial correlator according to an embodiment of the present invention. FIG. 6 is a diagram showing another expression of the carrier frequency phase candidate vector in FIG. 5. FIG. 4 is a diagram illustrating how a code index and a code phase according to an embodiment of the present invention are defined for a received signal. FIG. 6 is an enlarged view for illustrating how a code index and a code phase are defined for a received signal according to an embodiment of the present invention. FIG. 6 is a diagram illustrating how a modified code vector according to an embodiment of the present invention is generated based on an original code vector. FIG. 4 is a diagram illustrating a three-dimensional graph for code rate-phase candidate vectors to be used as the next correlator according to an embodiment of the present invention. FIG. 6 is a diagram illustrating how a set of code sequence start pointers according to the present invention are assigned to a modified code vector. FIG. 6 is a diagram illustrating how a data word according to an embodiment of the present invention is formed from a series of incoming level discrete sample values. FIG. 13 is a diagram illustrating how the data words of FIG. 12 are associated with various pre-generated vectors according to an embodiment of the present invention. FIG. 4 is a diagram illustrating how multiple pairs of pointers are assigned according to an embodiment of the present invention. It is a figure which shows the signal receiver which concerns on a proposal. FIG. 3 is a flow diagram illustrating a schematic method for processing a spread spectrum signal according to the present invention.

  In order to implement a truly software-based signal processing solution in real time at a high sampling rate, as required in GNSS, the architecture of the microprocessor system must be utilized very efficiently. The present invention uses an approach that includes a form of parallel processing based on a pre-generated code vector representation. Thereby, the present invention balances between processing speed and memory usage, which compensates for the performance deficiencies of general purpose microprocessors with respect to ASICs. Details of this solution will become apparent from the following disclosure.

FIG. 1 is a spectrum diagram for a representative spread spectrum signal from which information according to the present invention is extracted. For example, this signal may be a GPS C / A signal in the L1 band where the carrier frequency f _HF is 1.57542 GHz. This means that the spectrum of the signal is symmetric around the carrier frequency _fHF . Furthermore, this signal has at least 98% of its energy distributed within the 2 MHz wide frequency band BWHF near the carrier frequency f _HF . Thus, to allow further processing at lower frequencies, it is generally sufficient if the spread spectrum signal is downconverted to an intermediate frequency f _IF near 1 MHz.

FIG. 2 is a spectrum diagram of such a downconverted signal, the spectrum of which is symmetric around the intermediate frequency f _IF , which is considerably lower than the carrier frequency f _HF , for example f _IF = 1.25 MHz. . Movement of the transmitter (ie, typically an orbiting satellite) and the receiver (ie, typically a mobile GNSS receiver) causes a Doppler shift in the received signal that is Proportional to the relative speed to the receiver. Maximum Doppler shift _{f D} is defined, for a GNSS, usually about 10 kHz. Accordingly, the entire spectrum is reduced to an interval 2f _D so that the center frequency drops to an arbitrary position between f _IF + f _D (corresponding to the maximum positive Doppler shift) and f _IF −f _D (maximum negative Doppler shift). Shifted within. For illustration, the interval 2f _D is greatly exaggerated in FIG. The figure also shows a phase variation interval φ _D near the intermediate frequency f _IF (along the independent axis φ). The phase variation interval φ _D indicates the fact that the phase position of the received signal is initially unknown to the receiver and thus constitutes a calibration parameter. The effect of this parameter will be further elucidated below with reference to FIG.

An enlarged version of the spectrum in FIG. 2 is shown in FIG. Intermediate frequency, the center frequency is shifted to be in the interval 2f _D from _{f IF-min = f IF -f} D to _{f IF-max = f IF +} f D by possible Doppler effect. The potential maximum shifted spectrum is shown here with dotted lines. FIG. 3 also shows the Nyquist frequency r _S / 2 corresponding to the minimum sampling rate r _S , which, when used to sample the intermediate frequency signal, is a non-aliased discrete spectrum (non-aliased). discrete spectrum). If f _D = 10 kHz, f _IF = 1.25 MHz, BW _HF = 2 MHz and the intermediate frequency signal is low-pass filtered through an analog filter having a 2.5 MHz wide passband, the minimum sampling rate r _S Becomes 5 MHz (ie, twice the highest frequency component after passing through the analog filter).

  FIG. 4 schematically shows how the data signal D is modulated into a signal source specific code sequence CS at the transmitter side according to an embodiment of the present invention. According to a preferred embodiment of the present invention, the signal source specific code sequence CS constitutes so-called pseudo-random noise. Here, the data signal D includes a data symbol sequence [+1, -1, -1, +1, -1] and has a relatively low symbol rate of 50 Hz, for example. However, the source-specific code sequence CS has a relatively high symbol rate (or more precisely, a chipping rate). For example, a signal source-specific code sequence CS in the GPS C / A code format has a chipping rate of 1.023 MHz and includes 1023 chips for each period. Each chip ch is either +1 or -1. Thus, the C / A code repeats itself every ms. Data signal D is modulated into signal source-specific code sequence CS by multiplying each data symbol by signal source-specific code sequence CS (or spread by signal source-specific code sequence CS). Thereby, the data symbol +1 results in the code sequence CS without change, while the data symbol −1 results in the inverted code sequence CS.

  If a data signal D having a rate of 50 Hz is spread by a source-specific code sequence CS having a chipping rate of 1.023 MHz, this results in 20 complete code sequences CS per data symbol. . That is, the period time for one data symbol is 20 ms, while the period time for a code sequence is 1 ms.

FIG. 5 shows a three-dimensional graph for different carrier frequency phase candidate vectors. As described above, the intermediate frequency spectrum is Doppler shifted. Thus, the particular frequency of the received signal is initially unknown within the Doppler frequency interval from f _{IF- min} to f _IF-max . Furthermore, the phase position of the received signal is unknown to the receiver in advance. In order to enable the estimation of the actual intermediate carrier frequency, the Doppler frequency interval f _IF-min -f _IF-max is divided into an integer number of equidistant frequency steps. Of course, the higher the number of steps, the more accurately the intermediate carrier frequency can be estimated. In this example, a step size Δf of 20 Hz is selected. However, in some applications, a very large step size Δf provides sufficient frequency accuracy, while in other applications (which requires relatively high accuracy), a considerably smaller step size Δf is required. Is done. In any case, a 20 kHz Doppler frequency interval f _IF-min -f _IF-max and a step size Δf of 20 Hz requires 1000 steps, where the first step is from f _IF-min to f _IF includes frequencies up _-min + Delta] f, the second step _includes the frequencies from _{f IF-min} + Δf to _{f IF-min} + _2Δf, and so on to the last _{step, f IF-min} + _999Δf from _{f IF-} Includes frequencies up to _max .

A frequency candidate vector f _IF-C is defined for each frequency step. In order to estimate the initial phase position of the intermediate carrier frequency, an integer number of initial phase positions φ ₀ ,..., Φ ₇ are determined. The illustrated example shows such the eight positions phi _C. This gives a phase solution of π / 4. Accordingly, eight different phase positions φ ₀ ,..., Φ ₇ are defined for each frequency candidate vector f _IF-C , each representing a specific phase shift from 0 to 7π / 4. A combination of a certain carrier frequency candidate vector f _IF-C and a specific initial phase position φ _C is referred to as a carrier frequency phase candidate vector. FIG. 5 shows all such vectors (ie 8) for the first step, where f _IF-C = f _IF-min and for the second step f _IF-C = F _IF-min + Δf.

Further, FIG. 5 includes a shaft L _C, along its axis, the number of sample values of the frequency candidate vector f _IF-C (or length) is shown. Preferably, _{L C} number is a power of two. This is because it provides a simpler implementation. The number of sample values in each frequency candidate vector f _IF-C is set in consideration of an available memory space, a desired frequency solution, and a desired phase solution. If, in the above example, a length of 512 samples is selected and each sample requires 1 bit, the total (uncompressed) memory demand is 3.9 MB (ie 1000 frequencies x 8) Phase position × 512B = 4096000B = 3.9 MB).

FIG. 6 shows another expression of the carrier frequency phase candidate vector V _fφ (f _IF−C , φ _C ) in FIG. Here, the vector V _fφ (f _IF-C , φ _C ) is changed from the carrier frequency f _IF-C = f _IF-min (for example, 1.24 MHz) to f _IF-C = f _IF-max (for example, 1.26 MHz). and range up to, and the range of phase shift positions phi _C = phi ₀ from (e.g. 0) to φ _C = φ ₇ (e.g. 7 [pi] / 4), all, _{s n} samples of (e.g. 512) length It is represented by a cube (cube) and has a L _C. Thus, a particular frequency phase candidate vector V _fφ (f _IF−C , φ _C ) is given by a cube segment in a cube parallel to the L _C axis, where L _C = s ₁ to L _C = s _n Range.

After generating all the carrier frequency phase candidate vectors V _fφ (f _IF−C , φ _C ), the vectors can be easily accessed during subsequent acquisition and / or tracking phases. Stored in a digital memory having a relatively short access time. Preferably, the carrier frequency phase candidate vector V _fφ (f _IF-C , φ _C ) is an incoming data word (representing one or more received signals) and the carrier frequency phase candidate vector V _fφ (f _IF-C , Φ _C ) according to a data format suitable for performing a multiplication operation with the segment. This vector contains packed samples or is unpacked to the smallest manageable unit in the microprocessor, such as a byte. Details regarding further data format issues are discussed below.

FIG. 7 shows a graph for an exemplary code sequence CS, which has been received, down-converted, and sampled according to an embodiment of the present invention. After the subsequent removal of the intermediate frequency component, the code sequence CS acquires either a signal value of +1 or −1 and its symbol is modified, for example at a chipping rate of 1.023 MHz. On the receiver side, the code sequence CS is sampled at a sampling rate r _S of 5 MHz, for example. This sampling instance is indicated by a dot on the code sequence CS along the axis s _i . As is apparent from this figure, there are about 4 sample values for each chip section. Therefore, the requirements of the Nyquist theorem are clearly satisfied. However, the sample cannot determine the transition instance with sufficient accuracy, that is, the code sequence signal CS changes from one representative chip symbol to another representative chip symbol. Therefore, a so-called code phase must be determined. Furthermore, a specific sample value from which the code sequence starts must also be established.

  FIG. 8 shows a graph illustrating the sample values (so-called code index CI) where the code sequence CS starts and how the code phase Cph is defined with respect to this sample value. Therefore, the initial segment of the code sequence signal CS in FIG. 7 is represented in FIG.

Here, it is assumed that the first sample instance occurs shortly before the actual start of the signal section of the code sequence CS. Therefore, the corresponding sample value s ₁ belongs to the end point (end) of the preceding section. This is illustrated by the dashed portion of the line representing the code sequence CS. However, the subsequent sample instance overlaps the section of the code sequence CS. Corresponding sample value s ₂ is defined as the code index CI. The actual beginning of the code sequence signal is determined in the preceding acquisition phase and typically includes a correlation between the received signal and a partial copy of the code sequence CS.

Subsequently, the distance between the start of the code sequence signal CS and the code index CI is defined as the code phase Cph, and thus the translation (or skew) between the sample instance s _i and the chip transition. ) Measure. In order to estimate the code phase Cph, the sampling interval between two consecutive sample instances s _i is divided into an integer number of possible initial code phase positions. In this example, ten 0.1 steps from 0.0 to 0.9 are defined, where the first step 0.0 indicates that the code sequence signal CS starts with a code index CI, The last step 0.9 shows that the code sequence signal CS almost starts with the preceding sample value s ₁ . FIG. 8 shows the case where the code phase Cph is 0.4, ie the code sequence signal CS starts approximately in the middle of two consecutive sample instances s _i .

Similar to the carrier frequency phase candidate vector V _fφ (f _IF-C , φ _C ), the code sequence CS specific to each signal source and the combination of the code rate CR, the code index CI, and the code phase Cph are coded. A rate phase candidate vector may be defined. Nevertheless, according to a preferred embodiment of the present invention, a modified code vector is first generated based on each original code vector.

Figures 9 (a) and 9 (b) show how this is implemented. The code vector CV in FIG. 9A includes a large number of sample values, for example, 5000 sample values. Preferably, the number of sample values is selected in relation to the sampling rate so that a complete code sequence interval is recorded. In the above example, the sampling rate r _S is 5 MHz, which means that 5000 samples are exactly one complete code sequence CS section (it has a section of 1 ms and the chipping rate is 1023 MHz. To 1023 chips).

Here, the modified code vector CV _m is obtained by copying a specific number of elements E _e (for example, two) from the end of the original code vector CV to the beginning of the modified code vector CV _m. Generated based on vector CV. On the other hand, a specific number of elements E _b (for example, two) from the beginning of the original code vector CV are copied at the end of the code vector CV _m , that is, after the element E _e . FIG. 9 (b) shows the resulting modified code vector CV _m . In this example, each chip symbol is sampled with about 4 instances, so two sample values correspond to about half a chip interval. Thereby, the above copy of the elements E _e and E _b allows a one-half-chip shift of the code vector CV during the association period between the code vector CV and the received data. This is described in more detail below with reference to FIG.

FIG. 10 shows a three-dimensional graph relating to different code rate phase candidate vectors CV _m (CR _C-C , Cph) representing the code sequence CS (i) specific to the signal source. Similar to the intermediate carrier frequency, code rate CR _C for the nominal for code rate candidate vector _{CR C-C} may also vary due to the Doppler effect to the maximum value _{CR max} from the lowest value _{CR min.} However, since the code rate (or chipping rate) is relatively high, the rate variation here is rather small.

For example, a 10 kHz Doppler shift at the intermediate carrier frequency corresponds to a shift in the nominal code rate CRC at 1.023 MHz, which is only 6.5 Hz. That is, the Doppler shift for the code rate is the ratio between the carrier frequency and the code rate (1575.42 MHz / 1.023 MHz = 1540; thus, the 10 kHz Doppler shift at the intermediate carrier frequency is 10000 / 1540≈6. 5Hz). Code rate candidate vector _{CR C-C may} reach any of the values in the Doppler rate interval between CR min = 1022993.5Hz and _CR max = 102306.5Hz. The Doppler rate interval CR _min -CR _max is divided into an integer number of equidistant code rate steps, eg thirteen. In this example, therefore, each of the intervals CR _min , CR _min + ΔCR,..., CR _max spans 1 Hz. According to a preferred embodiment of the invention, the code rate step size is adaptive and is determined, for example, by the sampling frequency and the desired resolution.

In FIG. 10, the three-dimensional graph (or cube) includes a set of code rate phase candidate vectors CV _m (CR _C−C , Cph), where the code rate is from CR _C−C = CR _min to CR. Oyobi the range of _{C-C = CR max,} the code phase, range from Cph = 0.0 to a range of Cph = 0.9, and, for each vector _{s m} samples length _{L CVm} (e.g. 5004) Have Thus, a particular code rate phase candidate vector CV _m (CR _C−C , Cph) is given by a cubic segment CV _m (CR _min + 2ΔCR, 0.4) parallel to the L _CVm axis and L _CVm = It ranges from s ₁ to L _CVm = s _m .

After generating all code rate phase candidate vectors CV _m (CR _C-C , Cph), these vectors have relatively short access times so that they are easily accessed during the capture and / or tracking phase. Stored in digital memory. Preferably, the code rate phase candidate vector CV _m (CR _C-C , Cph) is implemented between vectors derived from the incoming data word and the code rate phase candidate vector CVm (CRC-C, Cph). Stored according to a data format compatible with the correlation operation. For example, the code rate phase candidate vector CV _m (CR _C-C , Cph) may include packed samples and may be unpacked to the smallest manageable unit in a microprocessor, such as a byte. Good. Details regarding further data format issues are discussed below with reference to FIG.

During the tracking of the spread spectrum signal, i.e. during the continuous reception of the signal and the demodulation of the data contained therein, it is necessary to update the corresponding tracking parameters for that signal. Usually, the tracking phase precedes the so-called acquisition phase, in which a set of preliminary parameters are established that are required to trigger signal decoding. Thus, if the acquisition phase is successful, at least one signal source specific code sequence is identified and the result is that the data signal transmitted by this sequence can be demodulated. The tracking characteristics associated with the source-specific code sequence include a modified code vector CV _m (defining carrier frequency candidate vector f _IF-C and initial phase position φ _C ), code phase position Cph, and code index. Including CI. In order to maintain the proper timing of the association operation at the receiver, a set of pointers are required, which indicate the starting position of the code sequence associated with the modified code vector and preferably each correlation. ) Is repeatedly updated.

FIG. 11 shows such a set of pointers assigned to the modified code vector CVm. Prompt pointer P _P indicates a current estimate of the code sequence start position. This pointer P _P is initially set equal to the code index CI. The appropriate pointer P _P for any next segment of received data is obtained by reading the appropriate modified code vector CV _m from the digital memory in which the pre-generated code vector is stored. Between each segment, not only the initial code phase Cph but also the code rate candidate vector CRC _-C is updated. Further, at least a pair of early pointer P _E, and late pointers P _L, respectively, assigned to each side of the prompt pointer P _P, where the early pointer P _E, the position of the prompt pointer P _P before at least one of identify the sample values are arranged in the elements, and, late pointer P _L specifies a sample value being positioned at least one element behind the location of the prompt pointer PP. Each correlation is implemented on the assumption that the code sequence starts not only at the P _L position, but also at the P _E and P _P positions. If association relates P _P position, a high correlation value than the correlation value regarding any other position brings (correlation value), if the balance is a P _E pointers and P _L pointer simultaneously (i.e., correlation of these positions This is interpreted as an optimal setting of the tracking parameters. However, this is not true, sampled data, as a correlation peak coincides with the position again PP pointer must be repositioned in relation to the P _P pointer and P _L pointers. Accordingly, the pointer arrangement is not calibrated (un-calibrated pointer positioning) is detected by the difference between the correlation value regarding the correlation value and the P _L pointer location about P _E pointer position while the association is still the highest correlation values for P _P positioned Bring. In this example, the potentially highest correlation value (corresponding to the optimal P _P positioning) is based on either the P _E pointer or the P _L pointer that yields the highest correlation value, and the current P _P position and the current P _E between position, or present somewhere between the current P _P position and the current P _L position. Typically, pointer updates occur between each segment. Further, a second pair or a third pair P _E pointers and P _L pointers outside the first pair of pointers, increasing the likelihood of achieving an accurate adjustment of the tracking. Based on the value of the received symbol, an optimal setting of the P _P pointer is effectively the equivalent association regarding P _P position result in lower correlation value than for any of the other pointer positions. That is, this is true if a negative data bit is received. However, otherwise the same principle applies.

In either case, P _P pointer P _E pointer is located to some within the interval of P _L pointer only modified code vector CV _m. P _P pointer may be set anywhere within a first interval 111, P _E pointer may be set to any position of the second interval 112, the P _L pointer optional third interval 113 It may be set at the position. All pointers are always set together so that their relative distance does not change. The limit value P _L-min for the P _L pointer matches the first element of the modified code vector CV _m . That is, outside this range, no significant association can be performed. As a result of this and the P _P , P _E , P _L pointers being set together, a corresponding limit value P _E-min for the P _E pointer is also defined.

FIG. 12 shows how data words d (1), d (2),..., D (N) are formed from an incoming level variance sample value series 1210 according to an embodiment of the present invention. . Data words d (1), d (2 ), ..., d (N) are formed such that each data word d (k) contains a number of elements equal to the number s _n of elements in each carrier frequency-phase candidate vector Is done. For example, the length of these vectors is 512 sample values (as in the example described above with reference to FIGS. 5 and 6). If each series 1210 contains 5000 sample values, the number of data words N is 10 (5000 / 512≈9.76 is the last quarter of the 10th data word d (10). Is empty). According to the invention, the elements in the current data word d (k) are processed together (ie in parallel). Then, the processing of the next data word d (k + 1) is started and the processing of the current data word d (k) required when processing the next data word d (k + 1) is being performed. Any processing parameters obtained in (1) are propagated as input data to subsequent processing. These processing parameters are preferably a pointer P _d indicating the first sample value of the next data word d (k + 1), a carrier frequency phase candidate vector V _fφ (f _IF−C , φ _C ) (ie phase (in-phase) and carrier frequency candidate vector f _IF-C representing the version, carrier frequency candidate vector f _IF-C representing a quadrature-phase (quadrature-phase) _version), the relevant set of code vectors CV _m, and A prompt pointer P _P , an early pointer P _E , and a late pointer P _L are included.

FIG. 13 illustrates how the data words of FIG. 12 according to embodiments of the present invention are associated with various pre-generated vectors. First, a corresponding set of carrier frequency phase candidate vectors V _fφ (f _IF−C , φ _C ) is calculated for each data word d (k). Typically, this set is obtained in the previous acquisition stage (or during the processing of the previous word if the data word d (k) is not the first word to be processed).

Then, for each carrier frequency phase candidate vector V _fφ (f _IF-C , φ _C ) in the related set, it is orthogonal to the previously generated in-phase representation f _{IFI of the} vector (V _fφ (f _IF-C , φ _C )). A phase representation f _IFQ is obtained respectively. Both representations can be found in a digital memory in which pre-generated carrier frequency phase candidate vectors V _fφ (f _IF-C , φ _C ) are stored. This is because they are simply the same vector delayed by a quarter interval, and the memory retains the initial phase position representation across all intervals. See FIG.

Then, in order to eliminate the influence of the carrier frequency component, the data word d (k) is multiplied by the in-phase representation f _IFI of the carrier frequency phase candidate vector V _fφ (f _IF−C , φ _C ) in the related set. As a result, the first intermediate frequency reduced information word _SIFI (k) is generated. In contrast, the data word d (k) is also multiplied by the quadrature phase representation f _IFQ of the carrier frequency phase candidate vector V _fφ (f _IF−C , φ _C ) in the associated set and reduced by the second intermediate frequency. Information word S _IF-Q (k) is generated.

Preferably, multiplication of each of the orthogonal phase representation f _IFQ and the in-phase representation f _IFI of the carrier frequency phase candidate vector V _fφ (f _IF−C , φ _C ) with the data word d (k) is performed by an XOR operation or a SIMD operation ( That is, each is implemented by 1-bit and multi-bit multiplication). That is, if the data word d (k) and the vector V _fφ (f _IF-C , φ _C ) have compatible data formats, this is possible.

The first intermediate frequency reduced information word S _IF-I (k) is then multiplied by the modified code vector CV _m-P , which has been read from the digital memory and its digital the memory, these pre generated vectors are stored, starting at a position supported by the prompt pointer P _P. A first prompt-despread symbol string Λ _IP (k) is generated as a result of this operation. The first intermediate frequency reduced information words S _{IF-I (k)} is multiplied with a modified code vector CV _{m-V (k),} which starts at the position indicated by the early pointer P _E . A first early-despread symbol string Λ _IE (k) is generated as a result of this operation. The first intermediate frequency reduced information word S _IF-I (k) is similarly multiplied by the modified code vector CV _m-L (k), which is at the position indicated by the late pointer P _L. Starting and a first late-despread symbol string Λ _IL (k) is generated. Furthermore, the first intermediate frequency reduced information word S _IF-Q (k) is multiplied by the modified code vector CV _m-P (k), which starts at the position indicated by the prompt pointer PP. And a second prompt despread symbol string Λ _QP (k) is generated. In contrast, the second intermediate frequency reduced information words S _{IF-Q (k)} is multiplied modified code vector CV _m-E and _(k), it was indicated by the early pointer P _E position And a second early despread symbol string Λ _QE (k) is generated. Similarly, the second intermediate frequency reduced information word S _IF-Q (k) is multiplied by the modified code vector CV _m-L (k), which is at the position indicated by the late pointer P _L. Starting and a second late despread symbol string Λ _QL (k) is generated.

Thereafter, the resulting data word _{D R-IE (k),} D R-IP (k), D R-IL (k), D R-QE (k), D R-QP (k), D R _-QL (k) is the despread symbol string Λ _IP (k), Λ _IE (k), Λ _IL (k), Λ _QP (k), Λ by adding together the elements in each string _It is derived from each of _QE (k) and Λ _QL (k). If expressed the string is not packed data, the processor, in order to obtain the resulting data word D _{R (k),} it may simply implement the relevant adding operation. However, if the string represents packed data, the resulting data word D _R (k) is in accordance with the preferred embodiment of the present invention and each despread symbol string Λ _IP (k) , Λ _IE (k), Λ _IL (k), Λ _QP (k), Λ _QE (k), Λ _QL (k) based on the bit patterns given by up).

According to a preferred embodiment of the invention, the individual data words D _R-IE (k), D _R-IP (k), D _R-IL (k), D _R-QE (k), D _R-QP (k), in addition to D _{R-QL (k),} after generating the data word D _{R (k),} so that the sum of all data words D _{R (1)} to D _{R (k)} is obtained, Corresponding accumulated data words are generated. Therefore, if the data word d (N) is processed, the resulting sum of DR (1) to DR (N) is also at hand.

Furthermore, the resulting data word represents payload information. For example, the resulting set of _DR-IE (k), DR _-IP (k), DR _-IL (k), _DR-QE (k), _{DR -QP} (k), D _{Based on R-QL} (k), a part of information of the demodulated transmission data symbol sequence D as shown in FIG. 4 is obtained. In steady state operation, basically only one of the resulting data words, eg DR _-IP (k), actually carries information.

According to embodiments of the present invention, each of the intermediate frequency reduced information words S _IF-I (k) and SIF-Q (k) and the modified code vectors CV _m-P (k), CV _{m- All} multiplications involving multiplication with _E (k); CV _m−L (k) are performed by XOR or SIMD. This is because the intermediate frequency information words S _IF-I (k); S _IF-Q (k) and the modified code vectors CV _m-E (k), CV _m-P (k); CV _m-L (k Yes, as long as they have compatible data formats.

It has been described with reference to FIG. 11 that early pointers and late pointer pairs are assigned to prompt pointers to achieve improved tracking. FIG. 14 shows that each of two pairs of such pointers P _E1 ; P _L1 and P _E2 ; P _L2 is assigned in the vicinity of the prompt pointer P _P according to an embodiment of the present invention. Here, the horizontal axis represents the code shift CS, and the vertical axis represents the normalized correlation factor. Therefore, the optimum position of the prompt pointer P _P is equal to the correlation value 1 normalized. The more the sampled data is shifted in either direction from this position, the lower the correlation value is substantially zero or more for one chip shift. Both the first pair of pointers P _E1 and P _L1 produce a first shifted correlation result (as shown) and both the second pair of pointers P _E2 and P _L2 are second. generating a shifted correlation result, the correlation first shifted correlation result is larger than the correlation result of the second shift, that and the second shifted correlation result obtained at the prompt pointer P _P If it is smaller than the result, the algorithm is perfectly balanced. Preferably, the correlation result is generated according to a process corresponding to the item described above with reference to FIG. Based on the correlation results for pointer positions P _P , P _E1 , P _L1 , P _E2 , and P _L2 (if necessary), modified code vectors CV _m-E (k), CV _m-P ( k), a set of CV _m-L (k) is selected and considered to be optimal.

  FIG. 15 shows a signal receiver 1500 for receiving navigation data signals transmitted in a navigation satellite system according to an embodiment of the present invention. The receiver 1500 includes a radio front end unit 1510, an interface unit 1520, and a digital processor unit 1530.

The wireless front end unit 1510 is configured to receive a continuous wireless signal S _HF and generate a relatively high frequency corresponding electrical signal S _IF in response thereto. The interface unit 1520 is configured to receive the electrical signal SIF and in response generate a series of sample values that represent the same information as the electrical signal SIF and are divided into data words d (k). Subsequently, the digital processor unit 1530 comprises a memory means, which is loaded with a computer program that, when run on the processor unit 1530, can control the steps of the proposed procedure. It is like that.

  For general purposes, a general method for processing a spread spectrum signal according to the present invention will now be described with reference to the flow diagram of FIG.

  The preparation step 1600 pre-generates code vectors, each of which is intended to be received (or at least capable of being received) and demodulated at the receiver. Represents a series. Step 1600 is performed before signal reception is activated.

  In Step 1610, a continuous signal having a relatively high frequency is received. In the following step 1620, the continuous signal is sampled at the basic sampling rate, thereby producing a resulting series of time discrete signal samples. Each sample is also quantized (with a relatively low resolution, such as 1 bit per sample value, or delivered from the wireless front-end unit 1510 so that a corresponding level-discrete sample is obtained. Quantized with a relatively high resolution depending on the number of data bits per sample taken and the application). Subsequently, in step 1630, data words are generated from the sample values, where each data word includes one or more consecutive sample values. Thereafter, in step 1640, correlation operations between information in the data word and at least one pre-generated representation of the source-specific code sequence are performed. This associating step associates each vector in at least a subgroup of code vectors with at least one vector obtained from a data word. In the following step 1650, data is generated as a result of the association performed in step 1640. Step 1660 then checks to see if the sampled sequence has been terminated, i.e., there are more data words to process. If additional sampling data to be processed is found, the procedure returns to step 1640. If not found, the procedure loops back to step 1610 again for continuous reception of incoming signals. Of course, such a signal is also preferably received during the execution of steps 1620 to 1660. The continuous procedure illustrated in FIG. 16 is merely applied to a specific received signal segment. In practice, all steps are 1610 to 1660 and are actually performed in parallel.

  Not only the sub-sequence of steps described with reference to FIG. 16 above, but also the processing steps may be controlled by a programmed computer device, such as a microprocessor provided in the GNSS receiver. Furthermore, although the embodiments of the present invention described with reference to the drawings include a computer apparatus and processing performed by the computer apparatus, the present invention is also adapted to implement a computer program, and in particular, the present invention. Extends to computer programs on or in the carrier. The program may be in the form of code intermediate source and object code, such as source code, object code, partially compiled form, or any other suitable for use in performing processing according to the present invention. It may be in the form of The carrier may be any entity or device that can carry the program. For example, the carrier may include a recording medium such as a CD (Compact Disc) or a ROM (Read Only Memory) such as a semiconductor ROM, or a magnetic recording medium such as a floppy (registered trademark) disk or a hard disk. Further, the carrier may be a transmissible carrier such as an electrical or optical signal transmitted via electrical or optical cable, wirelessly, or by other means. If the program is embodied in a signal carried directly by a cable or other device or means, the carrier is composed of such a cable or device or means. Alternatively, the carrier may be an integrated circuit in which the program is embedded, and the integrated circuit is configured to perform the associated process or is adapted for use in performing the associated process.

  As used herein, the term “comprises / comprising” is used to identify a specified feature, integer, step, or components. Used for. However, this term does not exclude the presence or addition of one or more additional features, integers, steps, or components or groups thereof.

  The present invention is not limited to the embodiments shown in the drawings, and can be freely modified within the scope of the claims of the present invention.

1500; signal receiver 1510; wireless front end unit 1520; interface unit 1530; digital processor unit

Claims (34)

A signal receiver (1500) for receiving and decoding a spread spectrum signal, comprising:
A wireless front end unit (1510) configured to receive a continuous radio signal (S _HF ) and generate a corresponding electrical signal (S _IF ) of a relatively high frequency in response to the continuous radio signal;
An interface unit (1520) configured to receive the electrical signal (S _IF ) and generate a series of sample values divided into data words (d (k)) in response to the electrical signal;
A plurality of decode vectors are pre-generated, the data word (d (k)) is input, and at least one of the decode vectors in the sub-group of the decode vectors is multiplied by at least one vector obtained from the input data word. A signal receiver comprising a digital processor unit (1530) configured to demodulate the signal by performing an association operation comprising.   The signal receiver of claim 1, wherein the digital processor unit is configured to multiply the received data word by a decode vector by multiplying a plurality of samples in parallel.   The signal receiver according to claim 2, wherein the digital processor unit is configured to multiply the plurality of samples in parallel by an XOR operation.   The signal receiver according to claim 2, wherein the digital processor unit is configured to multiply the plurality of samples in parallel by SIMD operation. The digital processor unit is configured to pre-generate the plurality of decode vectors such that one or more of the decode vectors are code vectors (CV; CV _m ), and the code vectors (CV; CV _m) is, represents a particular signal source specific code sequence (CS (i)), the coding sequence of claims 1 to 4 in which sampled and quantized by the basic sampling rate (r _s) The signal receiver according to any one of the preceding claims.   The signal receiver according to claim 5, wherein the signal source-specific code sequence (CS) is pseudo-random noise.   7. The digital processor unit according to claim 1, wherein the digital processor unit is configured to pre-generate the plurality of decode vectors such that one or more of the decode vectors are carrier frequency phase candidate vectors. The signal receiver described. The wireless front end unit is configured to down-convert an incoming high frequency signal (S _HF ) into an intermediate frequency signal (S _IF ), the high frequency signal (S _HF ) being a first frequency (F _HF ) has a symmetric spectrum, and the intermediate frequency signal (S _IF ) is symmetric near the second frequency (f _IF ) that is relatively lower than the first frequency (f _HF ). The signal receiver according to claim 1, wherein the signal receiver has a unique spectrum. The digital processor unit is:
Determining a maximum frequency variation (f _D ) of the second frequency (f _IF ) due to the Doppler effect;
A Doppler frequency interval (f _IF−min −f _IF−max ) is defined in the vicinity of the second frequency (f _IF ),
Dividing the Doppler frequency interval (f _IF−min −f _IF−max ) into an integer number of equidistant (Δf) frequency steps (f _IF−min −f _IF−min + Δf,..., F _IF−max );
By defining a frequency candidate vector (f _IF-C ) corresponding to each frequency step (f _IF−min −f _IF−min + Δf,..., F _IF−max ),
Configured to generate one or more carrier frequency candidate vectors;
The minimum frequency limit (f _IF-min ) in which the Doppler frequency interval (f _IF−min −f _IF−max ) is equal to the difference between the second frequency (f _IF ) and the maximum frequency variation (f _D ). And a maximum frequency limit (f _IF-max ) equal to the sum of the second frequency (f _IF ) and the maximum frequency variation (f _D ). The digital processor unit is:
Determining an integer number of initial phase positions (φ ₀ ,..., Φ ₇ ) for the frequency candidate vector (f _IF-C );
Determining a carrier frequency phase candidate vector (V _fφ (f _IF-C , φ _C )) for each combination of the carrier frequency candidate vector (f _IF-C ) and the initial phase position (φ ₀ ,..., Φ ₇ ). By
The signal receiver of claim 9, configured to generate one or more carrier frequency phase candidate vectors. The digital processor unit is:
Determining the number of elements (s _n ) in each carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C ));
The carrier frequency phase candidate according to a data format adapted to perform a multiplication operation between the data word (d (k)) and the carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) The signal receiver according to claim 10, configured to store a vector (V _fφ (f _IF−C , φ _C )). The digital processor unit _{obtains a} number of elements in which a segment of the carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) is equal to the number of elements in the data word (d (k)). Therefore, a segment of the carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) and one of the at least one vector (S _IF-I (k); S _IF-Q (k)) To each segment of the carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) so that one can be processed together by at least one of a SIMD operation and an XOR operation. The signal receiver of claim 11, wherein the signal receiver is configured to add at least one element. The digital processor unit is:
Determine the maximum variable code rate due to the Doppler effect,
A Doppler rate interval (CR _C-min -CR _C-max ) is determined near the center code rate (CR _C );
The Doppler rate interval (CR _C-min -CR _C-max ) is divided into an integer number of equidistant (ΔCR) code rate steps (CR _C-min , CRC _-min + ΔCR, ..., CRC _-max ). ,
By defining a code rate candidate (CR _C-C ) for each code rate step (CR _C-min , CRC _-min + ΔCR,..., CRC _-max ),
Configured to generate one or more code rate candidates,
The Doppler frequency interval includes a minimum code rate limit (CR _C-min ) equal to a difference between the center code rate (CR _C ) and the maximum code rate variation (CR _D ), and the center code rate (CR _C And a maximum code rate limit (CR _C-max ) equal to the sum of the maximum code rate variation (CR _D ). The digital processor unit is:
Determine an integer number of possible initial code phase positions (0.0,..., 0.9) for each code rate candidate (CR _C-C );
For each signal source-specific code sequence (CS (i)), a set of combinations between a code rate candidate (CR _C-C ) and a code phase position (0.0,..., 0.9) is defined. By
The signal receiver of claim 13, configured to generate one or more code rate phase candidate vectors. The digital processor unit is:
About the code sequence (CS (i)) specific to each signal source,
By generating a code vector (CV) corresponding to sampling each code rate-phase candidate vector with the basic sampling rate (r _s), claim configured to generate a set of code vectors (CV) 14. The signal receiver according to 14. The digital processor unit is:
Copying a specific number of elements (E _e ) from the end of the original code vector (CV) to the beginning of the modified code vector (CV _m );
By copying the element (E _b ) from the beginning of the original code vector (CV) to the end of the modified code vector (CV _m ),
The signal receiver according to claim 15, configured to generate a modified code vector (CV _m ) based on each code vector (CV). The digital processor unit is configured to store a set of modified code sequences ({CV _m (CR _C-C , Cph)}) for each signal source specific code sequence (CS (i)). ,
Where each modified code vector (CV _m ) includes a number of elements (s ₁ ,..., S _m ) representing a sampled version of at least one complete code sequence (CS);
Particular modified code vector (CV _m), the code rate candidate _{(CR C-C)} and code phase position (Cph) and signal receiver according to claim 16, wherein the defined for each combination of. The digital processor unit includes a modified code vector (CV _m ) and one of at least one vector (S _IF-I (k); S _IF-Q (k)) as a SIMD operation and an XOR operation. Of at least one vector (S _IF-I (k); S _IF-Q (k)) obtained from the data word (d (k)) so that they are processed together by at least one of The signal receiver of claim 17, wherein the signal receiver is configured to adapt a data format of the modified code vector (CV _m ) with respect to a data format.   19. The signal receiver according to claim 1, wherein the signal receiver is configured to receive a spread spectrum signal and decode the spread spectrum signal by a two-stage process including an initial supplementation step and a subsequent tracking step. Signal receiver.   The digital processor unit is a set required to trigger decoding of signals received during the subsequent tracking phase when the signal receiver is performing the initial supplemental phase. The signal receiver of claim 19, wherein the signal receiver is configured to establish a preliminary parameter of   The digital processor unit includes a source-specific code sequence (CS) used to generate a signal received by the signal receiver when the signal receiver is performing the initial supplementary stage. The signal receiver of claim 20, wherein the signal receiver is configured to identify The digital processor unit is configured to associate a set of preliminary parameters with the source-specific code sequence when the signal receiver is performing the initial supplementary phase, the preliminary parameters being
A modified code vector (CV _m );
A carrier frequency candidate vector (f _IF-C ),
Initial phase position (φ _C );
Code phase position (Cph);
Signal receiver of any one of claims 20 or 21 and a code index (CI) indicating the start sample value for the modified code vector (CV _m). The digital processor unit, when the signal receiver is performing the subsequent tracking step;
Calculate a prompt pointer (P _P ) for each modified code vector (CV _m ) based on the tracking characteristics;
Near each prompt pointer (P _P ) is configured to assign at least one pair of early and late pointers (P _E , P _L ; P _E1 , P _L1 ; P _E2 , P _L2 ),
The prompt pointer (P _P ) indicates a code sequence start position, and the initial prompt pointer (P _P ) is equal to the code index (CI);
The early pointer (P _E ) identifies a sample value positioned at least one element before the prompt pointer (P _P ) position;
23. A signal receiver as claimed in any one of claims 19 to 22, wherein the late pointer (P _L ) identifies a sample value located at least one element behind the prompt pointer (P _P ) position. The digital processor unit, when the signal receiver is performing the subsequent tracking step;
Receive an incoming series of level variance sample values (1210);
Said sample such that each data word (d (k)) contains a number of elements equal to the number of elements (s _n ) in each carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) Generates a data word (d (1),..., D (N)) of value (1210),
For the data word (d (k)), a corresponding set of carrier frequency phase candidate vectors (V _fφ (f _IF−C , φ _C )) is calculated,
For each carrier frequency phase candidate vector (V _fφ (f _IF-C , φ _C )) in the corresponding set, a pre-generated in-phase representation (f _IFI ) of the vector (V _fφ (f _IF-C , φ _C )). ₂₄ ) and a quadrature phase representation (f _IFQ ) respectively. _24. A signal receiver as claimed in any one of claims 19 to 23. The digital processor unit, when the signal receiver is performing the subsequent tracking step;
A first intermediate is obtained by multiplying each data word (d (k)) by the in _- phase representation (f _IFI ) of the carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) in the corresponding set. Generating a frequency-reduced information word (S _IF-I (k));
Each data word (d (k)) is multiplied by the quadrature phase representation (f _IFQ ) of the carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) in the corresponding set to obtain a second 25. A signal receiver according to claim 24, configured to generate an intermediate frequency reduced information word ( _SIF-Q (k)). The digital processor unit, when the signal receiver is performing the subsequent tracking step;
A modified code vector (CV _m-P (k)) starting at the position indicated by the prompt pointer (P _P ) and the first intermediate frequency reduced information word (S _IF-I (k)) To generate a first prompt despread symbol string (Λ _IP (k)),
A modified code vector (CV _m-E (k)) starting at the position indicated by the early pointer (P _E ) and the first intermediate frequency reduced information word (S _IF-I (k)) To produce a first early despread symbol string (Λ _IE (k)),
A modified code vector (CV _m-L (k)) starting at the position indicated by the late pointer (P _L ) and the first intermediate frequency reduced information word (S _IF-I (k)) To produce a first late despread symbol string (Λ _IL (k)),
A modified code vector (CV _m-P (k)) starting at the position indicated by the prompt pointer (P _P ) and the second intermediate frequency reduced information word (S _IF-Q (k)) To generate a second prompt despread symbol string (Λ _QP (k)),
A modified code vector (CV _m-E (k)) starting at the position indicated by the early pointer (P _E ) and the second intermediate frequency reduced information word (S _IF-Q (k)) To generate a second early despread symbol string (Λ _QE (k)),
A modified code vector (CV _m-L (k)) starting at the position indicated by the late pointer (P _L ) and the second intermediate frequency reduced information word (S _IF-Q (k)) 26. The signal receiver of claim 25, wherein the signal receiver is configured to multiply to generate a second late despread symbol string (Λ _QL (k)). The digital processor unit generates, for each despread symbol string, the resulting data word (D _R-IP (k), D _R-IE (k), D _R-IL (k), D _R-QP 27. The signal receiver of claim 26, configured to obtain (k), _DR-QE (k); _{DR -QL} (k)). The digital processor unit searches each pre-generated value in the table (1310) to obtain the resulting data word (D _R-IP (k), D _R-IE (k), D _R- 28. The signal receiver of claim 27, configured to obtain _IL (k), _DR-QP (k), _DR-QE (k); _{DR -QL} (k)). The digital processor unit performs an in _- phase representation (f _IFI ) of the carrier frequency phase candidate vector (V _fφ (f _IF-C , φ _C )) and the data word (at least one of a SIMD operation and an XOR operation). d (k)) and between the quadrature phase representation (f _IFQ ) of the carrier frequency phase candidate vector (V _fφ (f _IF−C , φ _C )) and the data word (d (k)). 29. A signal receiver according to any one of claims 24 to 28, configured to perform each of said multiplications. The digital processor unit includes the modified code vector (CV _m-P , CV _m-E ; CV _m-L ) and the intermediate frequency reduced information word by at least one of a SIMD operation and an XOR operation. _30. A signal receiver according to any one of claims 24 to 29, configured to perform the multiplication with ( _SIF-I (k); _SIF-Q (k)). The digital processor unit is:
In connection with completion of processing of the current data word (d (k)) and activation of processing of the next data word (d (k + 1)),
A pointer (P _d ) indicating the first sample value of the next data word (d (k + 1));
A group of parameters describing the relevant set of carrier frequency phase candidate vectors (V _fφ (f _IF−C , φ _C ));
The corresponding set of code vectors (CV _m );
31. A signal receiver as claimed in any one of claims 24 to 30, configured to propagate a prompt pointer, an early pointer, and a late pointer (P _P , P _E , P _L ). Receiving a relatively high frequency continuous signal (S _IF );
Sampling the continuous signal (S _IF ) at a basic sampling rate to generate a resulting series of time discrete signal samples (S [s _i ]);
Quantizing each signal sample into a corresponding level discrete sample value;
Generating a plurality of data words (d (1),..., D (N)) each including one or more consecutive sample values (s ₁ ,..., S _n );
A method of processing a spread spectrum signal comprising the step of associating information in said data word (d (k)) with at least one decoding vector,
The method includes a preparation for the association step, wherein a plurality of decode vectors are pre-generated before receiving the continuous signal (S _IF ),
The associating step includes multiplying each decoded vector in at least a subgroup of the decoded vectors by at least one vector obtained from the data word (d (k)).   33. A computer program that can be loaded directly into an internal memory of a computer and includes software for controlling the steps according to claim 32 when the program is run on the computer.   A computer-readable recording medium comprising a program recorded on the recording medium, wherein the program is for causing a computer to control the steps according to claim 32.

Images